I asked a question on EL&U which was taken upon negatively. It is linked here: “Is "tell advice" not idiomatic over "give advice"?

Let's look up the definition of "idiomatic" from the OED:

(1a) Relating to or exhibiting the forms of expression, grammatical constructions, phrases, etc. used in a distinctive way in a particular language, dialect, or language variety, formerly especially those considered nonstandard or colloquial. Now usually spec.: established by usage as having a meaning not deducible from the meanings of the individual words. https://www.oed.com/view/Entry/91033?redirectedFrom=Idiomatic#eid>

From my research people did not accept Google books/websites/etc. to be a good reflection of ‘idiomacy’, so I consulted Corpus of Contemporary American English (COCA) where “tell advice” is being used.

I was challenging a statement that it was not used by native speakers at all. This is extremely misleading. It’s like saying Standard English is the best so we will ignore all other dialects. A person notes that this could be “hyperbolic”, that people do but only a small minority, “microscopic in fact”. However to me, exaggeration doesn’t mask a false truth.

Here are my arguments:

  • The OED definition mentions nothing about native speakers. So perhaps people should just focus on less on whether a person is a native speaker, but on the wider sense of “idiomatic”. How is idiomacy tested? Is it impossible for a non-native but fluent speaker to speak idiomatically?

  • My evidence does show that native speakers are using from COCA. Google also showed 50,000 cases of where “tell advice” is used. Of course there are some questionable cases. It’s almost impossible to validate every source, however I feel like among the mass of 50,000 that it would be wrong to predict that there is not a single native speaker out there. What is undetermined is the actual % of 50,000 cases is “tell advice” being used by a native speaker.

  • Some dictionary definitions of idiomatic mention “native speakers” but I feel this is too simplistic of a definition as mentioned above.

  • If only a minority of native speakers is only using it, does that make it negligible or removed as an option at all? Because people are alluding to the latter, completely disregarding the alternative and justifying their use of “give advice” only on the basis that it is used more, not considering other options; even though “tell advice” violates no grammatical rule.

  • I would say: It is idiomatic in the community, that “tell advice” is being used. Perhaps not idiomatic in Standard English, but still in a community of native speakers or dialect - however small attesting its use. I think this fits in line with OED’s definition:

[...] Relating to or exhibiting the forms of expression, grammatical constructions, phrases, etc. used in a distinctive way in a particular language, dialect, or language variety [...]

Here is what CJ Dennis said in chat, which I complete agree with, particularly the second paragraph. :

There could arise some population of native English speakers for whom "tell advice" is idiomatic. It could become idiomatic in every English variety. However, it isn't at present and there's no evidence it ever was. "Long time no see" is ungrammatical but idiomatic, and therefore "good" English.

So there's no argument from a grammatical perspective to proscribe the use of a particular phrase, only the descriptive approach.

What baffles me is that people failed to see the merit of my question and just downvoted. If you look at the question I acknowledged from the get go that “give advice” was the most common variant, yet people focused more on disputing the evidence against “tell advice” as opposed to “give advice” saying it’s not natural and unidiomatic. COCA says otherwise.

I don’t understand why a perfectly good question is being treated like this. People are voting to close with no good basis or proper reason other than to incite that they do not like the usage of “tell advice”.

Clearly, it is a question well within the scopes of “English Language & Usage” but it’s not being given the merit of what it deserves.

Why do I have a reputation change on my reputation page that says "voting corrected"?

When a single user continually votes (up or down) on many of your posts within a short period of time, the system considers these votes to be invalid and removes them. This could happen for a variety of reasons, such as a user finding a user's great answer and visiting all of their posts to upvote them, or a user getting into an argument with another user and downvoting their posts indiscriminately in revenge. No matter the cause, this sort of systematic targeted voting is not considered normal behavior and the system will not allow it.

If such a voting pattern continues to happen between two users mutually or from one user towards another, or otherwise falls outside of normal voting patterns, moderators and/or developers may investigate the matter; intentionally voting merely to reduce or inflate another user's reputation is considered abuse.

Such votes will generally be invalidated as part of an automated process that runs every day, but may also be invalidated manually by the staff after an investigation. When the votes are invalidated, the reputation gain or loss from the votes is undone, which results in a record in the recipient's reputation history labelled "voting corrected".


I have had instances where my other answers have been downvoted simply because they did not like the question I asked above or that I was 'right'. If the down-vote are persistent and the down-votes are from the same group people - that can be construed as voting abuse.

Voting up a question or answer signals to the rest of the community that a post is interesting, well-researched, and useful, while voting down a post signals the opposite: that the post contains wrong information, is poorly researched, or fails to communicate information. The more that people vote on a post, the more certain future visitors can be of the quality of information contained within that post – not to mention that upvotes are a great way to thank the author of a good post for the time and effort put into writing it!

My post doesn't contain wrong information, is not poorly researched and does not fail to communicate information.

Even if it was to contain 'misinformation'* which is subjective, a user argues here:

I agree the wording "extreme cases" is very wrong here. A post providing misinformation is not an extreme case, it happens every day. That section in the help center wants to point out you can use down-votes as last resort when helping improving through comments doesn't help, and that is a good thing.

Contradictory guidelines for downvoting

*My question does not 'misinform' anyone. I say that "give advice" is the most common usage and that the grammatical 'tell advice' is just as valid as its counterpart and I have used evidence to support this.

The help section says:

When you vote down, you are nudging that content "down" the page, so it will be seen by fewer people. Voting down answers is not something we want you to take lightly, so it is not free.


What are the alternatives to downvoting? The upvote privilege comes first because that's what you should focus on: pushing great content to the top. Downvoting should be reserved for extreme cases. It's not meant as a substitute for communication and editing.

Instead of voting down:

  • If the post is spammy or offensive, flag it.
  • If the question is duplicate or off-topic, flag it for moderator attention.
  • If something is wrong, please leave a comment or edit the post to correct it.


  • english.stackexchange.com/q/456438/44619
    – Mari-Lou A
    Commented Feb 17, 2020 at 7:44
  • 2
    This is how voting works: people vote for the answers they think are correct, the ones they wish the next person visiting the question should see first and with the greatest weight. On EL&U, the majority of voters are also native speakers, and votes reflect what these native speakers believe native speakers say; it is possible to convince them of a usage they have not encountered naturally, but the bar for such arguments is raised significantly. Bottom line: your answer is downvoted because it conflicts with what natives speakers know to be true, & doesn’t meet that higher bar of convincing.
    – Dan Bron
    Commented Feb 17, 2020 at 13:30
  • @Dan Bron, it’s not about my answer, it’s about the question posed.
    – aesking
    Commented Feb 17, 2020 at 13:33
  • 3
    @aesking I think the same logic applies: the OP in that question appears to be arguing a position, a position at odds with what the voters believe to be true. This is the risk of asking a question with an embedded argument. There’s a kind of fuzzy gray zone between demonstrating you’ve done your research before asking your Q, and taking a position in it. (BTW, your meta-Q here brought my attention to that thread, I’d not seen it before and I still haven’t voted on the Q or any of the As, so take my comments here as speculating on the voting of others.)
    – Dan Bron
    Commented Feb 17, 2020 at 13:41
  • @Dan Bron would you agree it seems that it's not about the "quality" of the question in play here, even the question I asked on meta is receiving downvotes from the same people that downvoted my question! I wonder why? What I mean is the site rule is to use down vote "as a last resort". I believe someone also asked on meta: Why does EL&U gives less upvotes?. I think it's a bit unfair to use the downvote just because "they can" rather anything inherently to do with the question.
    – aesking
    Commented Feb 17, 2020 at 13:44
  • 3
    @aesking I don’t think there’s a site rule that “voting is a last resort”, and if there is it’s never been applied. Voting is intended to capture those subjective, hard-to-quantify qualities that the machine can’t; that’s why we have humans in the loop. The idea behind voting is precisely to let humans express “this is the kind of content I want on the site” or the opposite. We are encouraged to vote frequently to capture these preferences, to feed the machine, to build the kind of site we want to see. The very idea is to vote “just because we can” to capture those preferences.
    – Dan Bron
    Commented Feb 17, 2020 at 13:50
  • @Dan Bron "doesn't the guidance also say that down votes are supposed to be a last resort ?" - see this answer to Why do ELU users upvote so little and what can we do to fix this?
    – aesking
    Commented Feb 17, 2020 at 13:53
  • 2
    @aesking That user says the guidance says to use downvotes as a last resort, but I don’t know what guidance he’s referring to and I haven’t seen it (he doesn’t quote the help files, he just asserts that’s what they say). But no matter: as I say, even if the help files say that (I don’t think they do but they might), that is not how votes are or ever have been used. The “why” of that I put in my previous comment.
    – Dan Bron
    Commented Feb 17, 2020 at 13:58
  • @DanBron also site usage would disagree with you on that one, if the down-vote is presistent and the down-votes are from the same group people - that can be construed as voting abuse (see above and requires moderation attention). You displayed this behaviour when you said to me 'I will upvote everyone else's answer but not yours because my interpreation is that you have a 'chip on your shoulder' and basically that 'I find you annyoing'. Which has nothing to do with the quality of the question or answer posed.
    – aesking
    Commented Feb 17, 2020 at 17:33
  • 5
    1. Voting abuse happens when a single user downvotes a series of posts, usually within minutes of each other, by a different user. Having five downvotes on a single post means five individuals disagreed or disliked your post for whatever reason. On the other hand, I upvoted because I don't think the question is at all bad, it's perfectly on-topic for EL&U. I haven't upvoted here.
    – Mari-Lou A
    Commented Feb 17, 2020 at 17:49
  • @Mari-LouA +1 I agree, but if the same group of people that downvoted on there, also down-vote on here, what do you think? Also on EL&U, not on the meta, I noticed my reputation had gone down - some of my answers have been downvoted by what I think some suspect users who didn't like the question I posed above. This meta question is also receiving down-votes - it's not a bad question and people are really not following guidelines and using it as 'last resort'.
    – aesking
    Commented Feb 17, 2020 at 17:53
  • 2. The link to the question actually directs the visitor/viewer to the comment section. @aesking That letter in The Times was written by a Swedish national, Sven-Göran Eriksson, who had spent the first 34 years of his life in Sweden... so it's easy to understand why Dan Bron got confused.
    – Mari-Lou A
    Commented Feb 17, 2020 at 17:53
  • 4
    3. accusing users who downvoted your contribution as abusing the system, won't win you any friends. Take it from me, who's been here longer than you have. I'd rephrase the title if I were you.
    – Mari-Lou A
    Commented Feb 17, 2020 at 17:55
  • 1
    4. no, he made comments on my answer, which I've now converted into my question and hence deleted. The comment, which is the first one, i.e ** It is linked here**, was posted by CJ Dennis, UNDER *user067531's answer (oops, not your question. My bad). You have to admit, the link makes it confusing. P.S I also searched for @Shoe's comments but could not find them.
    – Mari-Lou A
    Commented Feb 17, 2020 at 18:03
  • 10
    Mari-Lou is right in all regards. With respect to voting abuse, it's no use bringing it up in a meta-question, there's nothing anyone but the mods or CMs can do. If you suspect voting abuse, you have 3 options, in order of increasing urgency: (1) wait 24 hours to see if the voting abuse script reverses any votes, (2) custom flag one of your posts for mod attention saying you suspect voting abuse and ask them to investigate, or (3) do the same, but using the "contact us" link in the footer of the page to contact the CMs (i.e. StackExchange employees) directly. Then it's up to them from there
    – Dan Bron
    Commented Feb 17, 2020 at 18:04

2 Answers 2


The OP, @aesking, told the community:

To me “to give someone advice” is the most common variant but it sounds phrasally and unidiomatic. However, to be politically correct I think “tell some advice” is more appropriate as personally in my grammar you cannot “give X advice”, you can only give physical objects but you can tell or speak advice/words.

What about “to give information”? What about “to give an opinion”? What about “to give an answer”? Isn't any response to a request or question, written or spoken, supplying that which was asked?

The OP's rational and justification is theirs to follow but they have yet to prove that their idiolect or that of other native speakers, tell someone advice, is the most natural and common in the English-speaking world, i.e. idiomatic.


I've given an answer to the OP, which I hope will clarify further. Unfortunately, due to the number of downvotes, their post no longer appears on the homepage. And my upvote has not helped to budge it out of oblivion; I'm also curious to see whether more users agree with the other answers posted, which are all basically saying that "tell someone advice" is uncommon in idiomatic speech. Not one answer has claimed it to be ungrammatical and I seriously doubt that @Shoe, as noted by @aesking, claimed otherwise.

  • 2
    P.S What does "it sounds phrasally" mean?
    – Mari-Lou A
    Commented Feb 17, 2020 at 11:10
  • 3
    The long thread that was subsequently moved to chat under the original post has this comment from Greybeard: "To add to the above: John told Stephanie some great advice. ->The collocation of "to tell advice" is incorrect. You can give/offer/proffer, etc., advice but you cannot tell advice." My comment, in response to a subsequent comment by aesking, part of which you reproduce above, contained the words: ... while to tell advice may not be considered ungrammatical, it is almost certainly not a collocation that a native speaker would use.
    – Shoe
    Commented Feb 17, 2020 at 12:52
  • 3
    On reflection, seeing the amount of sound and fury this issue has generated, it would have been preferable to comment: 'tell advice' is not ungrammatical but the more common collocation by far is 'give advice'.
    – Shoe
    Commented Feb 17, 2020 at 12:52
  • 1
    @Shoe - that’s what I’d been trying to do soon after the question was posted but the OP....refused to listen...
    – user 66974
    Commented Feb 17, 2020 at 13:02
  • @Shoe I did mention that in my post several times, but people seem to have ignored it!
    – aesking
    Commented Feb 17, 2020 at 13:34
  • 2
    Well, I upvoted your answer to that question even though the guy who asked it seems to have a chip on his shoulder. Now that I’ve seen the long chat where OP clearly demonstrates he doesn’t want answers, he wants validation of his point of view, I believe the downvotes on his Q are due to the shoulder chip more than anything else.
    – Dan Bron
    Commented Feb 17, 2020 at 14:17
  • @DanBron Hardly! You decided to comment on my answer to dispute, again the use of 'tell advice' rather than answering my question. Didn't you read that comments have been locked on that question for a reason? I will flag the unkind comment above and your other comments on there. So you saw my question here on meta and decided to light fuel to the fire rather than do anything constructive than to 'want to start a debate', this can be construed as harassment!. I will flag that comment. You said: it's not this collocation when my question states it's not a fixed collocation.!
    – aesking
    Commented Feb 17, 2020 at 14:35
  • 3
    @aesking You know, until now I really did not realize you were the same person asking the original Q, the answer you and I chatted under, and this meta-Q. I thought there were two people: the original question asker and you, who had answered it and asked this meta-Q. I was reading your long chats with other people under your original Q and it really seems to me you didn’t want to be convinced, you wanted to make an argument and have it validated, and resisted any counterpoints others raised. Feel free to flag any of my comments; it doesn’t bother me.
    – Dan Bron
    Commented Feb 17, 2020 at 14:41
  • @DanBron the reason that I wanted to validate my argument was for the sole purpose to make my question be taken seriously! It wasn't about about being right. It's not about personal opinions but actual theory/evidence. I've already purpoted that it is spoken by native speakers yet people are still focusing on that aspect rather than answering my question at all. 'Linguistics' as CJ Dennis said is 'science': you share views/give evidence and you dispute the evidence. That's all I did. But just because I disputed yours you will attack me! Find better evidence.
    – aesking
    Commented Feb 17, 2020 at 14:44
  • 3
    @aesking I did not attack you: I am telling you what I legitimately took away from reading your interactions under your Q. And I agree with the idea of giving and disputing evidence, and the advice to “find better evidence”. That’s precisely what I did under your answer. But now this thread is getting long and it seems heated, so I’ll leave it there.
    – Dan Bron
    Commented Feb 17, 2020 at 14:46
  • @DanBron How could you not know we were the same people, the first line of my question here literally says "I asked a question" // Well, frankly 'your evidence' or 'claims' were not relevant to my question and I disputed them. Though they were good points. Let's leave it at that. Good day sir!
    – aesking
    Commented Feb 17, 2020 at 14:55

The post starts off OK. "I was told X is true. I have found evidence to support the opposite, that ¬X is true."

However, instead of ending with "Is my argument that ¬X is true reasonable (therefore invalidating X)?" you end with "Please back me up and support my opinion that ¬X is true". (Note this is paraphrasing and not to be taken literally.) Your question was asked unscientifically. It did have evidence, but that evidence was very low quality (being based on Google search results). The subsequent edits have not shown substantial support for your position.

Science often requires the "5 sigma" standard, which is a less than 1 in 3.5 million chance for the raw data to have occurred from random fluctuations. While some evidence exists that some native speakers have occasionally uttered "tell advice", there's no evidence that any native speakers use "tell advice" in preference to "give advice" or that "tell advice" is used by any speaker on a frequent enough basis for it to be considered idiomatic. However, the support for "give advice" (which you asked it to be compared to) is undeniable.

"tell advice" could simply be a speaker saying "Let me tell you something advice". At the start of the sentence they planned to say "something" but changed their mind after they had already said "tell". Is that plausible enough to throw doubt on that evidence? Yes. Is it guaranteed to be true? No. More study would be needed to separate accidental use of "tell advice" with intentional use of "tell advice". It can't be proved either that it was intentional or accidental, so we don't know whether we can include that evidence or not. What we do know is that all humans make mistakes, and even the best speaker occasionally says the wrong word.

There is not enough quality evidence to reasonably support the position that "tell advice" is idiomatic.

You say yourself in this Meta question:

Here is what CJ Dennis said in chat, which I complete agree with [...]:

There could arise some population of native English speakers for whom "tell advice" is idiomatic. It could become idiomatic in every English variety. However, it isn't at present and there's no evidence it ever was.

[emphasis added]

You say you completely agree with my statement "['tell advice'] isn't [idiomatic] at present and there's no evidence it ever was." So I don't know why you keep arguing the opposite position.

Your question has been taken seriously by everyone who posted answers, comments, or edited your question. However, you have rejected every reasonable argument against your opinion. A question that only exists to support an unreasonable position is by definition a bad one and deserves downvotes (not you, the OP, just the question itself). Votes let users who have never seen the post before see how valuable other users have judged it to be so they can make a decision whether to read it or not.

I would ask that you don't take this personally. I can tell you're very emotionally invested in this question. Had it been asked from an enquiring position with a mind open to the opposite position, it might have been received more favourably. I would also suggest that you take some time away from English.SE to cool down.

I've had several long (multiple hour) chats with you recently. You seem like a nice person. It's just that on this issue you won't budge. You really need to have an open mind to see the value in the responses you got. You need to be able to judge the evidence objectively, and not from the viewpoint of an assumed conclusion. Refusing to give up a bad assumption is bad science and makes for a bad question. Not all evidence is equally valuable. It can also be very difficult when you are very emotionally invested in the assumption. If you backed down and admitted you were wrong you might feel very foolish, however, I don't think many people (if any) would think less of you for that. I for one would not judge you on that. I would admire you for overcoming your emotions.

The issue is not so much about whether you think X or ¬X is true, it's about whether a random reasonable person with no prior exposure to the question would judge X or ¬X to be true.

I really hope this post has helped you understand the dynamics of voting (that for most users it's based on evidence and presenting a reasonable argument) even though I haven't explained voting that much.

Good luck, and all the best.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .